On Jul 11, 2005, at 12:52 PM, Aaron Mulder wrote:
On Mon, 11 Jul 2005, Geir Magnusson Jr. wrote:
Yep - I would suggest then we keep it simple and have
/geronimmo/sandbox
Fine with me
/geronimo/sandbox/misc/SoC
I think we agreed this one is going to operate through patches.
Wasn't sure what we agreed on. If patches, great. If working as
"regular" committer, great.
/geronimo/sandbox/donations/trifork
/geronimo/sandbox/donations/ibm
I would prefer a mixed name or feature name rather than a company
name -- so perhaps trifork-corba or corba, and web-console or
ibm-web-console, or something like that.
Fine by me. I figured that TriFork *and* IBM might be contributing
more than one thing, so was a neat bucketing. But i don't care.
I'd be happy w/ separate ACLs to let people work as fast and
"normally" as possible, w/o having to wait for patches to be
accepted. There's no danger with SVN. That said, I'd go w/ patches
if that was the consensus.
Well, I don't want to offend anyone, but I can envision a scenario
where we don't see eye to eye with a submitter on architecture or
features
or whatever.
Heresy! Has it EVER been the case where we don't all harmoniously
agree? I'm shocked! SHOCKED!
:D
LOL
If the submitter charges ahead with their own changes in
their own style and that turns out to be unacceptable to us, then the
whole module is wasted.
The rules of technical consensus apply as for all our code - any one
of us can veto for technical reasons. This isn't "Geroni-forge" :)
If they submit patches instead, we are free to
accept them or hold off and massage the code into something more
appropriate to us. I am not saying this is the expected case,
which is
why I only have a minor preference for patches, but it would be
nice to
account for.
Right - all I'm saying is that anyone can veto a change and roll it
back out of SVN, so the end result is identical - community oversight
and control of the technical nature and changes to our codebase.
geir
Aaron
--
Geir Magnusson Jr +1-203-665-6437
[EMAIL PROTECTED]
